Accounting for the effect of forest and fragmentation in probabilistic rockfall hazard

Lanfranconi, Camilla; Frattini, Paolo; Sala, Gianluca; Bertolo, Davide; Sun, Juanjuan; Crosta, Giovanni Battista

doi:https://doi.org/10.5194/egusphere-2022-1521

Preprints

https://doi.org/10.5194/egusphere-2022-1521

Preprints

07 Feb 2023

| 07 Feb 2023

Accounting for the effect of forest and fragmentation in probabilistic rockfall hazard

Camilla Lanfranconi, Paolo Frattini, Gianluca Sala, Davide Bertolo, Juanjuan Sun, and Giovanni Battista Crosta

Abstract. The presence of trees along the slope and block fragmentation at impact strongly affect rockfall dynamics, and hazard as a consequence. However, these phenomena are rarely simulated explicitly in rockfall studies. We performed rockfall simulations by using the 3D rockfall simulator Hy-Stone, modelling both the presence of trees and fragmentation through specific algorithms implemented in the code. By comparing these simulations with a more classical approach that attempt to account implicitly for such phenomena in the model parameters, and by using a new probabilistic rockfall hazard analysis (PRHA) method, we were able to quantify the impact of these phenomena on the design of countermeasures and on hazard.

We demonstrate that hazard changes significantly when accounting explicitly for these phenomena, and that a classical implicit approach usually overestimates both the hazard level and the 95^th percentile of kinetic energy, leading to an oversizing of mitigation measures.

Received: 31 Dec 2022 – Discussion started: 07 Feb 2023

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this preprint. The responsibility to include appropriate place names lies with the authors.

Download & links

Preprint (PDF, 1700 KB)

Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
Preprint (1700 KB)

Supplement (1630 KB)

Download & links

The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.

Journal article(s) based on this preprint

28 Jun 2023

Accounting for the effect of forest and fragmentation in probabilistic rockfall hazard

Camilla Lanfranconi, Paolo Frattini, Gianluca Sala, Giuseppe Dattola, Davide Bertolo, Juanjuan Sun, and Giovanni Battista Crosta

Nat. Hazards Earth Syst. Sci., 23, 2349–2363, https://doi.org/10.5194/nhess-23-2349-2023,https://doi.org/10.5194/nhess-23-2349-2023, 2023

Short summary

Camilla Lanfranconi, Paolo Frattini, Gianluca Sala, Davide Bertolo, Juanjuan Sun, and Giovanni Battista Crosta

Interactive discussion

Status: closed

RC1:
'Comment on egusphere-2022-1521', Anonymous Referee #1, 06 Mar 2023

This manuscript is very interesting and worth to be published. The effect of fragmentation and forest interaction in rockfall hazard estimation are hot topics on the subject, and this contribution elegantly clarifies and quantifies the effect of both phenomena using the well-established rockfall simulation code Hy-Stone.
I suggest a few improvements before publication.
First, and regarding the fragmentation algorithm described in section 2.1.3, I suggest a more in detail explanation of each section of the algorithm. Even though some details on how fragmentation works on Hy-Stone can be found in Frattini et al. (2012), some parts of the process are not clear. Being the quantification of fragmentation effect one of the goals of this article I suggest a more technical description of the algorithm, including an explanation of how the Yashima et al. (1987) criterion is adapted for the fragmentation threshold and the criteria used to determine the post fragmentation directions (since this may control the spreading of the model).
Second, in the analysis of Roisan event, where fragmentation played a significant role as detailed in the manuscript, it would be very interesting to show the volume distribution of the deposited fragments in the field and compare it with the volumetric distribution obtained by the model during the calibration. If the model results correspond to the field measurements, the model calibration would be more complete regarding fragmentation. In this sense, it would also be convenient to explain what initial volumes are simulated to represent the event and how that initial distribution was determined. These suggestions could be included in section 3.1 if appropriate.
Some minor comments:
Pag 5 line 133. For the calculation of the transit relative frequency when considering fragmentation, the total number of simulated paths include all generated fragments during propagation? When no fragmentation occurs this calculation looks trivial, but increasing the number of blocks during the propagation may lead to misuse of the classic approach.
Pag 7 line 214: Typo: «Figure 2Figure 2»
Pag 12 line 359: Is the sentence «From the results of hazard computation with and without fragmentation algorithm, which the hazard decreases.» correct? Can’t get the meaning.
Pag 13 line 391: Possible bullet missing.
Congratulations for this outstanding work.

Citation: https://doi.org/10.5194/egusphere-2022-1521-RC1
- AC1: 'Reply on RC1', Camilla Lanfranconi, 18 Apr 2023
  
  Reply to Anonymous Referee #1
  
  First, and regarding the fragmentation algorithm described in section 2.1.3, I suggest a more in detail
  
  explanation of each section of the algorithm. Even though some details on how fragmentation works on Hy-
  
  Stone can be found in Frattini et al. (2012), some parts of the process are not clear. Being the quantification
  
  of fragmentation effect one of the goals of this article I suggest a more technical description of the algorithm,
  
  including an explanation of how the Yashima et al. (1987) criterion is adapted for the fragmentation threshold
  
  and the criteria used to determine the post fragmentation directions (since this may control the spreading of
  
  the model).
  
  Thank you for your comment. In the revised version, we will add a more detailed section regarding
  
  the Hy-Stone fragmentation algorithm and the Yashima et al. (1987) criterion adopted.
  Second, in the analysis of Roisan event, where fragmentation played a significant role as detailed in the
  
  manuscript, it would be very interesting to show the volume distribution of the deposited fragments in the
  
  field and compare it with the volumetric distribution obtained by the model during the calibration. If the model
  
  results correspond to the field measurements, the model calibration would be more complete regarding
  
  fragmentation. In this sense, it would also be convenient to explain what initial volumes are simulated to
  
  represent the event and how that initial distribution was determined. These suggestions could be included in
  
  section 3.1 if appropriate.
  
  Thank you for your suggestions. For the revised paper, we will prepare a figure comparing the
  
  volumetric distributions of the in-situ block size distribution (IBSD), the mapped block size
  
  distribution in the deposit (RBSD), and the simulated block size distribution with and without the
  
  algorithm. The figure will be added to the supplementary materials. Additionally, we will included a
  
  brief explanation in section 3.1 about the initial volumes simulated and how they were identified.
  Some minor comments:
  
  Pag 5 line 133. For the calculation of the transit relative frequency when considering fragmentation, the total
  
  number of simulated paths include all generated fragments during propagation? When no fragmentation
  
  occurs this calculation looks trivial, but increasing the number of blocks during the propagation may lead to
  
  misuse of the classic approach.
  
  Indeed, the comment is very interesting. In the calculation of the transit frequency, we use the
  
  number of simulated block before fragmentation. This may see counterintuitive because
  
  fragmentation can generate many blocks passing through a cell, and the value of frequency can
  
  potentially be larger than 1 (when the number of blocks passing through a cell is larger than the
  
  number of simulated blocks). This is correct, because this value is multiplied by the onset frequency,
  
  which is calculated without considering the dynamic fragmentation, and therefore it is consistent
  
  with the number of simulated blocks.
  Pag 7 line 214: Typo: «Figure 2Figure 2»
  
  Fixed
  Pag 12 line 359: Is the sentence «From the results of hazard computation with and without fragmentation
  
  algorithm, which the hazard decreases.» correct? Can’t get the meaning.
  
  The sentence is incorrect. It will be replaced with “The results with and without the fragmentation
  
  algorithm shows that the hazard level decreases when fragmentation is explicitly simulated”
  Pag 13 line 391: Possible bullet missing.
  
  Fixed
  
  Citation: https://doi.org/10.5194/egusphere-2022-1521-AC1
RC2:
'Comment on egusphere-2022-1521', Anonymous Referee #2, 25 Mar 2023

The manuscript represents a very interesting contribution to the understanding of the effect of forest and fragmentation in probabilistic 3D rockfall modelling. The rockfall model HY-STONE and the applied rockfall hazard methodology are valid and outlined clearly. However, the manuscript still lacks a bit of a clear structure, with in particular a sharp distinction between methods, results, and discussion. Furthermore, please check the required reference formatting and reference completeness text/list.
Some comments:
Line 80: please add if block volume & form are considered in Hy-STONE and how the forest/trees are included/considered in the model
Line 155: HY-STONE is capable of simulating such large volumes?
Line 214: typo “Figure 2Figure2”
line 286: typo "Rockfall Hazard"
Figures: please add the resolution of the data/simulations

Citation: https://doi.org/10.5194/egusphere-2022-1521-RC2
- AC2: 'Reply on RC2', Camilla Lanfranconi, 18 Apr 2023
  
  Reply to Anonymous Referee #2
  The manuscript represents a very interesting contribution to the understanding of the effect of forest and fragmentation in probabilistic 3D rockfall modelling. The rockfall model HY-STONE and the applied rockfall hazard methodology are valid and outlined clearly. However, the manuscript still lacks a bit of a clear structure, with in particular a sharp distinction between methods, results, and discussion. Furthermore,
  
  please check the required reference formatting and reference completeness text/list.
  
  In the revised version, we will attempt to make the chapter division more sharp. However, the overall paper is already divided into methods, analysis and results, and discussion.
  
  We will also check the references carefully.
  Some comments:
  
  Line 80: please add if block volume & form are considered in Hy-STONE and how the forest/trees are included/considered in the model
  
  Thank you for the comment. In the revised version, we will provide more details on how the tree impact algorithm works. The volume and shape of the blocks are taken into consideration by Hy- Stone. For both case studies, spherical blocks were simulated.
  Line 155: HY-STONE is capable of simulating such large volumes?
  
  Yes, Hy-Stone simulates individual blocks and it is able to compute the volumes described in the manuscript.
  Line 214: typo “Figure 2Figure2”
  
  Fixed
  line 286: typo "Rockfall Hazard"
  
  Fixed
  Figures: please add the resolution of the data/simulations
  
  As stated in line 172, we simulated all scenarios using the 1 x 1 m Lidar DTM of Aosta Valley Region. The parameters used with the two algorithms and the characteristics of each scenario are reported in supplementary Table S1, S2, and S3. In the revised version, we will add details about the simulated volumes and the number of simulated blocks.
  
  Citation: https://doi.org/10.5194/egusphere-2022-1521-AC2

Interactive discussion

Status: closed

RC1:
'Comment on egusphere-2022-1521', Anonymous Referee #1, 06 Mar 2023

This manuscript is very interesting and worth to be published. The effect of fragmentation and forest interaction in rockfall hazard estimation are hot topics on the subject, and this contribution elegantly clarifies and quantifies the effect of both phenomena using the well-established rockfall simulation code Hy-Stone.
I suggest a few improvements before publication.
First, and regarding the fragmentation algorithm described in section 2.1.3, I suggest a more in detail explanation of each section of the algorithm. Even though some details on how fragmentation works on Hy-Stone can be found in Frattini et al. (2012), some parts of the process are not clear. Being the quantification of fragmentation effect one of the goals of this article I suggest a more technical description of the algorithm, including an explanation of how the Yashima et al. (1987) criterion is adapted for the fragmentation threshold and the criteria used to determine the post fragmentation directions (since this may control the spreading of the model).
Second, in the analysis of Roisan event, where fragmentation played a significant role as detailed in the manuscript, it would be very interesting to show the volume distribution of the deposited fragments in the field and compare it with the volumetric distribution obtained by the model during the calibration. If the model results correspond to the field measurements, the model calibration would be more complete regarding fragmentation. In this sense, it would also be convenient to explain what initial volumes are simulated to represent the event and how that initial distribution was determined. These suggestions could be included in section 3.1 if appropriate.
Some minor comments:
Pag 5 line 133. For the calculation of the transit relative frequency when considering fragmentation, the total number of simulated paths include all generated fragments during propagation? When no fragmentation occurs this calculation looks trivial, but increasing the number of blocks during the propagation may lead to misuse of the classic approach.
Pag 7 line 214: Typo: «Figure 2Figure 2»
Pag 12 line 359: Is the sentence «From the results of hazard computation with and without fragmentation algorithm, which the hazard decreases.» correct? Can’t get the meaning.
Pag 13 line 391: Possible bullet missing.
Congratulations for this outstanding work.

Citation: https://doi.org/10.5194/egusphere-2022-1521-RC1
- AC1: 'Reply on RC1', Camilla Lanfranconi, 18 Apr 2023
  
  Reply to Anonymous Referee #1
  
  First, and regarding the fragmentation algorithm described in section 2.1.3, I suggest a more in detail
  
  explanation of each section of the algorithm. Even though some details on how fragmentation works on Hy-
  
  Stone can be found in Frattini et al. (2012), some parts of the process are not clear. Being the quantification
  
  of fragmentation effect one of the goals of this article I suggest a more technical description of the algorithm,
  
  including an explanation of how the Yashima et al. (1987) criterion is adapted for the fragmentation threshold
  
  and the criteria used to determine the post fragmentation directions (since this may control the spreading of
  
  the model).
  
  Thank you for your comment. In the revised version, we will add a more detailed section regarding
  
  the Hy-Stone fragmentation algorithm and the Yashima et al. (1987) criterion adopted.
  Second, in the analysis of Roisan event, where fragmentation played a significant role as detailed in the
  
  manuscript, it would be very interesting to show the volume distribution of the deposited fragments in the
  
  field and compare it with the volumetric distribution obtained by the model during the calibration. If the model
  
  results correspond to the field measurements, the model calibration would be more complete regarding
  
  fragmentation. In this sense, it would also be convenient to explain what initial volumes are simulated to
  
  represent the event and how that initial distribution was determined. These suggestions could be included in
  
  section 3.1 if appropriate.
  
  Thank you for your suggestions. For the revised paper, we will prepare a figure comparing the
  
  volumetric distributions of the in-situ block size distribution (IBSD), the mapped block size
  
  distribution in the deposit (RBSD), and the simulated block size distribution with and without the
  
  algorithm. The figure will be added to the supplementary materials. Additionally, we will included a
  
  brief explanation in section 3.1 about the initial volumes simulated and how they were identified.
  Some minor comments:
  
  Pag 5 line 133. For the calculation of the transit relative frequency when considering fragmentation, the total
  
  number of simulated paths include all generated fragments during propagation? When no fragmentation
  
  occurs this calculation looks trivial, but increasing the number of blocks during the propagation may lead to
  
  misuse of the classic approach.
  
  Indeed, the comment is very interesting. In the calculation of the transit frequency, we use the
  
  number of simulated block before fragmentation. This may see counterintuitive because
  
  fragmentation can generate many blocks passing through a cell, and the value of frequency can
  
  potentially be larger than 1 (when the number of blocks passing through a cell is larger than the
  
  number of simulated blocks). This is correct, because this value is multiplied by the onset frequency,
  
  which is calculated without considering the dynamic fragmentation, and therefore it is consistent
  
  with the number of simulated blocks.
  Pag 7 line 214: Typo: «Figure 2Figure 2»
  
  Fixed
  Pag 12 line 359: Is the sentence «From the results of hazard computation with and without fragmentation
  
  algorithm, which the hazard decreases.» correct? Can’t get the meaning.
  
  The sentence is incorrect. It will be replaced with “The results with and without the fragmentation
  
  algorithm shows that the hazard level decreases when fragmentation is explicitly simulated”
  Pag 13 line 391: Possible bullet missing.
  
  Fixed
  
  Citation: https://doi.org/10.5194/egusphere-2022-1521-AC1
RC2:
'Comment on egusphere-2022-1521', Anonymous Referee #2, 25 Mar 2023

The manuscript represents a very interesting contribution to the understanding of the effect of forest and fragmentation in probabilistic 3D rockfall modelling. The rockfall model HY-STONE and the applied rockfall hazard methodology are valid and outlined clearly. However, the manuscript still lacks a bit of a clear structure, with in particular a sharp distinction between methods, results, and discussion. Furthermore, please check the required reference formatting and reference completeness text/list.
Some comments:
Line 80: please add if block volume & form are considered in Hy-STONE and how the forest/trees are included/considered in the model
Line 155: HY-STONE is capable of simulating such large volumes?
Line 214: typo “Figure 2Figure2”
line 286: typo "Rockfall Hazard"
Figures: please add the resolution of the data/simulations

Citation: https://doi.org/10.5194/egusphere-2022-1521-RC2
- AC2: 'Reply on RC2', Camilla Lanfranconi, 18 Apr 2023
  
  Reply to Anonymous Referee #2
  The manuscript represents a very interesting contribution to the understanding of the effect of forest and fragmentation in probabilistic 3D rockfall modelling. The rockfall model HY-STONE and the applied rockfall hazard methodology are valid and outlined clearly. However, the manuscript still lacks a bit of a clear structure, with in particular a sharp distinction between methods, results, and discussion. Furthermore,
  
  please check the required reference formatting and reference completeness text/list.
  
  In the revised version, we will attempt to make the chapter division more sharp. However, the overall paper is already divided into methods, analysis and results, and discussion.
  
  We will also check the references carefully.
  Some comments:
  
  Line 80: please add if block volume & form are considered in Hy-STONE and how the forest/trees are included/considered in the model
  
  Thank you for the comment. In the revised version, we will provide more details on how the tree impact algorithm works. The volume and shape of the blocks are taken into consideration by Hy- Stone. For both case studies, spherical blocks were simulated.
  Line 155: HY-STONE is capable of simulating such large volumes?
  
  Yes, Hy-Stone simulates individual blocks and it is able to compute the volumes described in the manuscript.
  Line 214: typo “Figure 2Figure2”
  
  Fixed
  line 286: typo "Rockfall Hazard"
  
  Fixed
  Figures: please add the resolution of the data/simulations
  
  As stated in line 172, we simulated all scenarios using the 1 x 1 m Lidar DTM of Aosta Valley Region. The parameters used with the two algorithms and the characteristics of each scenario are reported in supplementary Table S1, S2, and S3. In the revised version, we will add details about the simulated volumes and the number of simulated blocks.
  
  Citation: https://doi.org/10.5194/egusphere-2022-1521-AC2

Peer review completion

AR: Author's response | RR: Referee report | ED: Editor decision | EF: Editorial file upload

ED: Publish subject to minor revisions (review by editor) (26 Apr 2023) by Andreas Günther

Dear Authors,
Referee reports on your paper indicate that only minor revisions would be required to accept your paper for publication in our journal. Reading your very interesting paper I agree with the reviewers that your paper must not necessarily be send out for another round of review. However, I made some additional observations listed below that you also should consider during revision. I very much agree with the major comments of referee #1 concerning simulation of rock fragmentation and calibration of the algorithm.
To proceed with the closed peer-review process, please submit your revised manuscript together with a point-per-point reply to the comments of the reviewers and the editor, and a manuscript version in track change mode indicating the applied changes. After resubmission, your paper will be reviewed again by the editor.
Best regards, Andreas

P2L32: „all relevant methodological issues“: what are these?
P2L43: “more rarely”: is this true given the large body of references cited here?
P4L100: Maybe a reference should be given that PRHA is initially based on the well-established concepts of PSHA?
P4L116: In Formula (5), shouldn’t F be replaced by f0?
P6L155: would be nice to have the coordinates also plotted in the maps in Fig.1
P7L196-201: shouldn’t rather Fig1C and Fig1D be referenced here?
P7L201: Table 1 and Table 2?
P9L253: A wall is not visible in Fig 4?
P9L255: I think it would be nice to leave at least one sentence in the text explaining the magnitude scenarios considered
P10L287-288 why and how?
P10L310: how was the combination done?
P10L310-L317 Check this paragraph some broken sentence
P11L331: is risk or its mitigation discussed anywhere in the paper?
P12L359-364: This is all not clear, please rephrase and explain

Hide

AR by Camilla Lanfranconi on behalf of the Authors (05 May 2023) Author's response Author's tracked changes Manuscript

ED: Publish as is (17 May 2023) by Andreas Günther

AR by Camilla Lanfranconi on behalf of the Authors (25 May 2023)

Journal article(s) based on this preprint

28 Jun 2023

Accounting for the effect of forest and fragmentation in probabilistic rockfall hazard

Camilla Lanfranconi, Paolo Frattini, Gianluca Sala, Giuseppe Dattola, Davide Bertolo, Juanjuan Sun, and Giovanni Battista Crosta

Nat. Hazards Earth Syst. Sci., 23, 2349–2363, https://doi.org/10.5194/nhess-23-2349-2023,https://doi.org/10.5194/nhess-23-2349-2023, 2023

Short summary

Camilla Lanfranconi, Paolo Frattini, Gianluca Sala, Davide Bertolo, Juanjuan Sun, and Giovanni Battista Crosta

Supplement

https://doi.org/10.5194/egusphere-2022-1521-supplement

Camilla Lanfranconi, Paolo Frattini, Gianluca Sala, Davide Bertolo, Juanjuan Sun, and Giovanni Battista Crosta

Viewed

Total article views: 306 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
210	80	16	306	33	2	2

HTML: 210
PDF: 80
XML: 16
Total: 306
Supplement: 33
BibTeX: 2
EndNote: 2

Views and downloads (calculated since 07 Feb 2023)

Month	HTML	PDF	XML	Total
Feb 2023	92	28	6	126
Mar 2023	51	19	4	74
Apr 2023	40	16	5	61
May 2023	11	9	0	20
Jun 2023	16	8	1	25
Jul 2023	0
Aug 2023	0
Sep 2023	0
Oct 2023	0
Nov 2023	0
Dec 2023	0
Jan 2024	0
Feb 2024	0
Mar 2024	0
Apr 2024	0
May 2024	0
Jun 2024	0
Jul 2024	0
Aug 2024	0
Sep 2024	0

Cumulative views and downloads (calculated since 07 Feb 2023)

Month	HTML	PDF	XML	Total
Feb 2023	92	28	6	126
Mar 2023	51	19	4	74
Apr 2023	40	16	5	61
May 2023	11	9	0	20
Jun 2023	16	8	1	25
Jul 2023	0
Aug 2023	0
Sep 2023	0
Oct 2023	0
Nov 2023	0
Dec 2023	0
Jan 2024	0
Feb 2024	0
Mar 2024	0
Apr 2024	0
May 2024	0
Jun 2024	0
Jul 2024	0
Aug 2024	0
Sep 2024	0

Viewed (geographical distribution)

Total article views: 302 (including HTML, PDF, and XML) Thereof 302 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 06 Sep 2024

Download

The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.

Preprint (1700 KB)
Metadata XML

Short summary

This paper presents a study on rockfall dynamics and hazard, examining the impact of the presence of trees along slope and block fragmentation. We compared rockfall simulations that explicitly model the presence of trees and fragmentation with a classical approach that account for these phenomena in model parameters. Both the hazard and the kinetic energy change. We also used a non-parametric probabilistic rockfall hazard analysis method for hazard mapping.


Total:	0
HTML:	0
PDF:	0
XML:	0